• Email
  • Print

Koestler’s Solution

Janus: A Summing Up

by Arthur Koestler
Random House, 354 pp., $10.00

Theodor Schwann, a founder of cell theory in the mid-nineteenth century, described life as “nothing but the form under which substances capable of imbibition crystallize.” This kind of reductionism did have a certain vogue in Schwann’s day, and its vestiges continue to color and constrain biological thought—most notably in the early and heady days of DNA, when some over-zealous biochemists thought they might render all of life’s complexity as a one-way readout of instructions coded on chromosomes.

Arthur Koestler believes that we stand at the brink of disaster, mired in “the sterile deserts of reductionist philosophy.” In this summing up of all his writings on science, Koestler has appointed himself irrigator of the deserts and has attempted to sow seeds, using his concept of “hierarchy.” Koestler identifies as two main obstructions to understanding first his own version of Darwinian theory (which is very different from Darwin’s or anything that a modern biologist would recognize); and second the penchant for causal determinism that closes our minds to psychic and acausal phenomena.

The structures of our world, Koestler argues, are ordered into hierarchies, or, as he calls them, “holarchies.” Each item, or “holon,” is both a self-contained entity and an element of the next level. Koestler has named his book for the double role played by each holon: “The holons are…Janus-like entities: the face turned toward the higher levels in the holarchy is that of a subordinate part in a larger system; the face turned toward the lower levels shows a quasi-autonomous whole in its own right.” The reductionist program is bankrupt because each level has both autonomy and its own unique contribution to make to the next level; the whole system cannot be reduced to molecules in motion at its base.

For Koestler the human dilemma is just one example of hierarchical ordering. As Janus-faced entities, we are caught between self-assertive tendencies appropriate to our level and integrative urges leading up toward the next.

We are faced with a contrast between the mature restraint of the self-assertive tendency and the immature vagaries of the integrative tendency…. Both the glory and the pathology of the human condition derive from our powers of self-transcendence, which are equally capable of turning us into artists, saints or killers, but more likely into killers.

As our loyalties are split between self and group, so too are our brains. The neocortex, seat of our intellect, has superimposed itself upon the older reptilian and mammalian brain. But—and this is our particular tragedy—we evolved so rapidly that the neocortex was unable to establish proper integration with and control over the primitive brain.

Homo sapiens may be an aberrant biological species, an evolutionary misfit, afflicted by an endemic disorder…. There is a flaw, some potentially fatal engineering error built into our native equipment…. This is the hideous but plausible hypothesis which any serious inquiry into man’s condition has to face.

Koestler cites symptoms of this disorder: infanticide and human sacrifice, our lack of inhibition toward killing members of our own species, the split between reason and emotion, and the failure of moral improvement to match technological advance. Arguing that “a biological malfunction needs a biological corrective,” Koestler proposes a biotechnical fix, a “combination of benevolent hormones or enzymes which would resolve the conflict between the old and recent structures in the brain, by providing the neocortex with the power of hierarchic control over the archaic lower centers, and thus catalyze the transition from maniac to man.” Perhaps, after giving our informed consent, we could add it to table salt and cure human nature as iodine once relieved goiters.

Quite a mouthful, but one never could accuse Koestler of narrow pedantry. I warmly endorse his general vision: a hierarchy of interacting levels seems a better metaphor for our complex world than a congeries of objects, each decomposable into its constituent atoms. But the utility of an abstract theme does not guarantee the truth of any set of particulars gathered under it. I regard Koestler’s “theory” of the human dilemma as singularly unconvincing and full of irony and contradiction. In carrying his scheme forward, he relies upon arguments rooted in the very habits of mind that he deplores at the outset. In the end, his book is a testament to what he attacks—prejudice imposed by conventional habits of Western thought. I rest my case on four paradoxes.

1) Koestler plays down the importance of new facts as motivating or requiring shifts in theory, a perspective I share with him. Yet, in some crucial places, his failure to consider new facts greatly compromises his formulation.

Koestler regards our capacity to kill one another as the primary evidence for our status as evolutionary misfits; for other species refrain from murdering their own and settle conflicts by ritual threat: “Man is alone (apart from some controversial phenomena among rats and ants) in practicing intra-specific murder on an individual and collective scale.” Koestler’s position is the one that Konrad Lorenz championed some twenty years ago and that achieved wide circulation. But it is wrong. We now know that infanticide, for example, is widespread among animals ranging from lions to langur monkeys.1 Moreover, the behavior is adaptive not pathological;2 human infanticide, though practiced for different reasons, is also widespread and usually adaptive (see Marvin Harris, Cannibals and Kings). I do not regard the infanticide of male lions as comparable in any way to the human case. I doubt that our capacity for infanticide arises from any direct genetic selection at all. Nonetheless, the widespread tendency among animals to kill members of their own species removes the label of unique pathology from our own practices and invalidates half of Koestler’s arguments for our status as evolutionary misfits.

Even worse, Koestler’s proposal for the cause (not only the symptoms) of our misbegotten evolution is based upon another error. He writes: “The neocortex of the hominids expanded in the last half a million years at an explosive speed which is without precedent in the history of evolution…. The result in this case seems to have been that the rapidly developing thinking cap, which endowed man with his reasoning powers, did not become properly integrated and coordinated with the ancient emotion-bound structures on which it was superimposed with such unprecedented speed.”

Again, Koestler repeats a claim often made a decade ago, although later evidence suggests that it is wrong. The last half million years witnessed the replacement of Homo erectus, with a brain of 1000 cc, by Homo sapiens, with our average of some 1400 cc. We have no evidence at all for an even transition between the two at a sharply accelerated rate of increasing brain size. Homo erectus lived for perhaps two million years and displayed no trend at all toward an enlarging brain. Homo sapiens appeared about 50,000 years ago with as large a brain as we have now. This is the standard paleontological pattern of abrupt replacement.3 It is common to nearly all transitions in the fossil record. The differences between Homo erectus and Homo sapiens are not inconsiderable, but the pattern of replacement is normal and the amount of change not unusual. The previous transition from Australopithecus africanus to Homo habilis may have involved a larger relative increase in brain size, and may have occurred just as rapidly.

2) Koestler devotes a large part of his book to ridiculing the theory of natural selection by presenting it as a caricature of itself. Yet this very theory may provide a resolution of his central dilemma by providing a mechanism for his perceptive metaphor.

Koestler caricatures the theory of natural selection as an egregious example of mechanistic reductionism. We evolutionists maintain, he thinks, that all organic change is fortuitous since selection can only eliminate the unfit; that organisms are passive entities buffeted about by shifting environments; and that genes are independent particles which must be concatenated one by one to produce any evolutionary transformation. But evolutionists view natural selection as a holistic theory based upon interactions between organism and environment.

First of all, the fortuitous variations that arise within the genetic makeup of organisms are only raw material. They direct nothing. Natural selection accumulates these variations and builds adaptation by steps along favored paths. Natural selection is a theory of creative, constructive change, not random alteration. Koestler attacks his nonexistent version of Darwinism by pointing to strong similarities between Australian marsupials and independently evolved “placentals,” i.e., most other mammals. “Why, if evolution were a free-for-all, did Australia not produce some entirely different species of animals?” But it did, and we call them kangaroos.

As for the similarities between kangaroos and other animals, they are striking, but not unexpected on the principle that well-constructed designs and ways of making a living are both limited. Since evolution is not a free-for-all (and no Darwinian ever said it was), we expect strong convergence between related animals living in similar niches. Koestler cites the “nearly identical skulls” of marsupial and placental carnivores as an evolutionary event beyond the power of natural selection. But the similarities reflect common shape and proportion for common function. The skulls are similar in adaptive design, not in structure; point by point, for all the standard differences between the two great groups, these skulls remain unambiguously marsupial and placental.

Secondly, Darwinians do not hold that organisms are passive agents in their own alteration. Darwinians, according to Koestler, maintain that behavior can change only as a result of fortuitously altered structure. He states that Lamarckians invoke, while Darwinians deny, the cardinal insight that organisms are active agents and that altered behavior may antedate and serve as an impetus for changes in morphology. But in fact Darwin agreed with Koestler, and so have Darwinians ever since, for altered behavior is an important mechanism for setting new selective pressures. Darwin wrote:4

It is difficult to tell, and immaterial for us, whether habits generally change first and structure afterward; or whether slight modifications of structure lead to changed habits…. The larger titmouse (Parus major) may be seen climbing branches like a creeper; it often, like a shrike, kills small birds by blows on the head; and I have many times seen and heard it hammering the seeds of the yew on a branch, and thus breaking them like a nuthatch.

Finally, Darwinians are not “genetic atomists,” but organic holists. Koestler writes: “Genetic atomism is dead…. The living organism is not a mosaic, where each bit is governed by a separate gene, and evolution does not proceed by replacing individual bits in a haphazard fashion until, lo and behold, the image of a fish is replaced by that of an amphibian.” Correct, but the scientists who killed such atomism were Darwinian naturalists who recognized the complexity of interaction between form and environment and who insisted that natural selection can only work upon whole organisms (in any case, genetic atomism never flourished beyond a few laboratories of experimental genetics).

Koestler might understand this better if he got his facts in order. He cites the extra “finger” of the giant panda, so useful for manipulating bamboo shoots, as an adaptation that could not be produced by fortuitous Darwinian variation. How could a structure so integrated with the rest of the panda’s body arise piece by piece and gene by gene? “The chances that among all possible genetic mutations just those which produced the added bones, nerves, muscles, and arteries occurred simultaneously and independently from each other are infinitesimally small.” Koestler thinks that the added finger is built of the same parts as all others and that these parts, if we accept the principle of natural selection, must have arisen de novo, one by one and somehow coordinated.

But he is wrong. The stunning fact about this extra “finger” is that while it functions as a finger it is not, anatomically, a finger at all. It is built from a detached sesamoid bone of the wrist. Darwin frequently cited such jerry-built structures as a major argument for natural selection: an organism must work with what it has. It cannot produce the best possible design by its internal will or the fiat of some external, directing force. In any case, super-numerary teeth and digits are not rare (polydactylism occurs in humans, for example); they are not built gene by gene but differentiated along with the others from a somewhat flexible, embryonic field.

If Koestler would look upon natural selection with more favor, I think he might come to see it as a potential mechanism for resolving what he sees as the central dilemma of human life. We are caught in a web of conflicting loyalties to ourselves and our groups. This tension between “self-assertive” and “integrative” tendencies is the cause of our most intense agonies, the source both of our creativity and of our greatest crimes—genocide in the name of higher principles. Koestler explains this conflict by linking it to the general role of objects in hierarchies—to their simultaneous existence as things in themselves and parts of higher levels. I regard this statement as both perceptive and enlightening; it may even be more than a metaphor, if hierarchy is the key to universal structure, as Koestler believes. But hierarchy is not a mechanism. Koestler provides no insight into causes when he states that we act as we do because we occupy a level in a hierarchy. Our capacity for behaving in different ways must have arisen somewhere and for some reason. My office and my pancreas are both part and whole, but they neither suffer nor act in confusion.

Could natural selection have set the possibilities for different kinds of behavior precisely because they are adaptive in hierarchies? The primary statement of natural selection, the “Darwinian imperative,” holds that individuals are selected to maximize the contribution of their own genes to future generations. We might think that animals follow this imperative by looking out for themselves at all times, and that only self-assertive tendencies could be favored by selection. But self-sacrifice for the good of a group can benefit ones own genes if the altruistic act saves a number of relatives who, in aggregate, bear more of an individual’s genes than the individual does himself. (Since my children share half my genes, I would increase my own Darwinian fitness by sacrificing myself for three of them.)

I separate myself sharply from the human sociobiologists who advocate direct genetic selection for specific human behaviors in the spectrum from self-assertive to integrative—xenophobia, conformity, aggression, etc. My view is exactly the opposite, though it arises from the same Darwinian theory of social behavior. I believe that selection has operated only to set the ranges of adaptive human behavior, not to specify definite acts in particular circumstances. The sociobiologist’s favorite theory of kin selection—the Darwinian justification for altruism—does not increase the scope of genetic determinism. Quite the contrary: it expands the range of behavior favored by selection to include Koestler’s integrative tendencies. In Civilization and Its Discontents Freud argued that only self-serving behavior can be biologically advantageous. In Freud’s view, complex society, which requires integration, forces us to act against our nature, thus making our entire civilization neurotic. But the capacity for integrative behavior may be as natural as the potential for self-assertive actions. Our dilemma lies in the fact that we are not programmed to exhibit one or the other in specific circumstances. Selection endowed us with the capacity for both, and we struggle in confusion on our intermediate rung of Koestler’s hierarchy.

3) Koestler, so fearless in smiting the windmills of convention, permits the most ancient Western prejudice of all to color his basic decision.

Koestler simply cannot face a world without inherent purpose. Of Darwinism, he writes: “It is a depressing doctrine, whether true or not. The indications are that it is not.” I have already argued that his indications are both misinformed and directed toward a caricature of the theory. Koestler is left with his hope of finding some purpose in nature. And he invests that hope in Lamarckism. He admits that Lamarckian experiments have failed, and the inheritance of acquired characters has not been demonstrated. But Darwinism, he argues, is in equally bad shape, so why not root for an intelligible world. “On the other hand, the patient labors of the Darwinian geneticists on thousands of generations of Drosophila have also failed to produce any evolutionary improvement. As far as the direct experimental evidence is concerned, the two sides might have called it quits.”

This is surely the most amazing statement in Koestler’s book. What does he think the geneticists who have been breeding Drosophila for sixty years have been doing? They have performed thousands of selection experiments and most have been successful; their experience with Drosophila is an illustration in the laboratory of realized evolution on Darwinian principles—while no one has ever performed a successful Lamarckian experiment. Drosophila labs have not made a superfly or transformed a fruit fly into a mammal, but they haven’t been trying to do so and would never have been so foolish. Evolution is not a tale of inexorable, cosmic improvement, but the story of adaptation to changing environments—and this kind of “improvement” occurs by selection over and over again in Drosophila.

Koestler molds an evolutionary theory to fit his hope for an inherently purposive, improving world. This is one of the deepest and oldest prejudices of Western thought. “The Purposer is each and every individual organism from the inception of life, which struggled and strove to make the best of its limited possibilities; and the sum total of these ontogenies reflects the active striving of living matter toward the optimal realization of the planet’s evolutionary potential.”

Lamarckism would indeed incarnate this reverie, but the world does not operate according to our hopes. If it did, we would still be created in God’s image and residing at the center of a limited universe. Darwin’s world does not incorporate this hope. It does present a vision of order and adaptation produced by the invisible hand of organisms seeking to maximize their own reproductive success, and nothing else. But what does this say about human purpose? Nothing that I can see. Why does Koestler seek so desperately to put purpose and improvement into the workings of nature? We ought to view Darwin’s message to us as an exhilarating one; for it tells us to construct purpose for ourselves and by our own decisions. We should not seek it in nature and hope to map ourselves passively upon it.

4) Most ironically of all, this passionate anti-reductionist uses (perhaps unbeknownst to him) the oldest reductionist argument about the brain as a basis for his theory of the human dilemma.

We are evolutionary misfits, Koestler maintains, because our neocortex, the locus of intellect, has not established proper control over our ancient brain, the locus of emotion. His argument depends upon the idea of a brain partitioned into its faculties by phylogenetic order:

The old brain got in the way of, or acted as a brake on, the new…. The marvelous potentialities of the unsolicited gift were only allowed to manifest themselves in the service of archaic, emotion-based beliefs…. The fault was not in our stars, but in the horse and crocodile which we carry inside our skulls.

No one would deny that the brain exhibits localization of function, but this does not make it a collection of independent units; and the functions, in any case, are not stacked one upon the other in evolution. The brain evolves by reorganization, not by addition.5 There is no thinking machine in the rind of our neocortex trying to exert its dominion over a dark world of animal emotion within.

Koestler’s simplistic regional model of the brain had its historical roots in phrenology, the early nineteenth-century science of mapping functions and emotions through bumps and depressions of the skull onto definite regions of the brain underneath. Phrenologists waged the campaign for their “science” with one persistent philosophical message—the validity of reductionism and its equation with true scientific procedure. They saw themselves as warriors for science against superstitions of an irreducible human spirituality that could not be located in any material structure. The richness of our thought and feeling can be decomposed into “faculties,” and each can be mapped, one for one, upon an atom of brain. The brain is a collection of independent parts. George Combe, a leading phrenologist, wrote in 1840:6

In a few years, it will appear as a singular fact in the history of the mind, that in the nineteenth century, men…should have seriously denied that the mind, in this world, acts by means of material organs…. If, then, the mind manifest a plurality of faculites, and if the brain be the organ of mind, it appears to be a sound inference that the brain may consist of a plurality of organs.

Brain research may be the major field in which Koestler’s general anti-reductionist stance needs firm assertion against lingering atomism. Patients were once lobotomized in the name of reductionism. Neurosurgeons finally realized that the prefrontal lobes are not the seat of specific disorders, but a substantial part of an integrated organ—but this came too late for tens of thousands. Several more recent proposals for psychosurgery continue to assert the idea of a decomposable brain—which is, as Koestler ought to be saying, reductionistic prejudice, not science. I am sorry that Koestler, to support his specific theory of human behavior, has cast his lot with the very perspective he rejects in his general vision.

  1. 1

    E.O. Wilson, Sociobiology (Harvard University Press, 1975), p. 85.

  2. 2

    In lions and langurs, males often kill young fathered by other males when they take over a female. Although it may appear both gruesome and senseless to us, these acts of infanticide increase the genetic fitness of child killers. Natural selection works to maximize the contribution of one’s own genes to future generations. New males have no genetic stake at all in the progeny of the males they displace; they will benefit themselves only by fathering and rearing their own brood. In human societies with well-established customs of infanticide, adaptive reasons differ but also confer increased fitness. Children are sacrificed when attempts to raise them would place such a strain on resources that the whole family (the entirety of a parent’s genetic stake in the next generation) would suffer.

  3. 3

    For an explanation of why this should be so and why it conforms with our best understanding of evolutionary theory, see S.J. Gould and N. Eldredge, “Punctuated equilibria: the tempo and mode of evolution reconsidered,” Paleobiology vol. 3, 1977, pp. 115-151. I recently wrote a short, general defense of punctuational change in the Sunday New York Times, Section 4, January 22, 1978.

  4. 4

    Origin of Species, first edition of 1859, pp. 183-184.

  5. 5

    Harry Jerison, The Evolution of the Brain and Intelligence (Academic Press, 1973).

  6. 6

    American Journal of Science, vol. 38, 1840, pp. 349-353.

  • Email
  • Print